6 Mission-critical info for uncrewed systems professionals Platform one Edinburgh engineers have developed computational and experimental tools to enable autonomous robots to maintain a steady position amid irregular, buffeting waves at sea, writes Nick Flaherty. Wave-detecting devices are tethered to the seafloor, and measure the direction and height of incoming waves, relaying the information in real time to a robot working nearby. This enables uncrewed machines to pre-empt complex waves, precisely counteracting these to maintain a stable position. This is suitable for autonomous underwater or surface vessels used for monitoring or for offshore wind-turbine maintenance. The non-linear, predictive, dynamic positioning model for remotely operated vehicles is compatible with robots operating both near the surface or at greater depths, where disturbances can still be felt strongly. The researchers say the technology improves on conventional control systems, which operate in a corrective fashion with a feedback loop. These are prone to respond slowly to the ocean’s fast-changing, nonlinear disturbances, making them much less reliable in such conditions. A major limitation is that a vessel needs to be able to perceive and counteract environmental disturbances effectively, restricting the use of small subsea vehicles, which can be tossed about in bad weather. “By forming a prediction of future wave disturbances and integrating this within the control system, we are able to expand this range with little to no change to the robot hardware. In terms of translating this technology into the field, this is a huge benefit and makes our system applicable to most vehicles on the market,” said Dr Kyle Walker, who developed the work as part of his PhD at the University of Edinburgh. The complete end-to-end control architecture combines a non-linear model predictive controller (NMPC) combined with a deterministic sea-wave predictor (DSWP). The DSWP continuously measures wave elevation upstream of the vehicle to form a forecast of its temporal evolution at the vehicle’s location. The predicted wave parameters are then used to estimate the impending wave-induced hydrodynamic loading, giving accurate short-term disturbance forecasts that can be used by the control system of a USV or UUV. Experimental testing confirmed the validity of the wave predictor, producing an error as low as 1.7 mm. The researchers said the NMPC outperforms a similar disturbance-mitigating feed-forward controller, with average improvements of up to 52%. It even performs better with noisy, lower accuracy wave predictions, and with communication time delays. Future research aims to improve the ability of autonomous machines to perform precise tasks, such as using robotic arms to detect rust or fix electric equipment while holding a steady position in water. Underwater Detectors enable robots to pre-empt complex waves at sea Novel algorithms for predicting waves (Image courtesy of Uncrewed Systems Technology) December/January 2025 | Uncrewed Systems Technology
RkJQdWJsaXNoZXIy MjI2Mzk4